Skip to content

Implement native orc input format without arrow to improve performance#53324

Merged
al13n321 merged 21 commits intoClickHouse:masterfrom
bigo-sg:ch_gluten_2583
Aug 21, 2023
Merged

Implement native orc input format without arrow to improve performance#53324
al13n321 merged 21 commits intoClickHouse:masterfrom
bigo-sg:ch_gluten_2583

Conversation

@taiyang-li
Copy link
Copy Markdown
Contributor

@taiyang-li taiyang-li commented Aug 11, 2023 •
edited
Loading

Changelog category (leave one):

  • Performance Improvement

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

implement native orc input format without arrow to improve performance

TODO

Q1: read only int type columns, speed up by 1.23x

select reporttime,appid,uid,clientversioncode,sdkversioncode,starttimestamp,backgroundtotal,foregroundtotal,firstiframesize,firstiframedecodetime,extras,mclientip,mnc,mcc,vsipsuccess,msipsuccess,vsipfail,msipfail,mediaflag,videomutetype,owneruid from file('./bigolive_audience_stats_orc.orc', NativeORC)    Format Null; 

0 rows in set. Elapsed: 0.329 sec. Processed 1.20 million rows, 227.43 MB (3.65 million rows/s., 690.48 MB/s.)
0 rows in set. Elapsed: 0.351 sec. Processed 1.20 million rows, 227.43 MB (3.43 million rows/s., 648.51 MB/s.)
0 rows in set. Elapsed: 0.359 sec. Processed 1.20 million rows, 227.43 MB (3.35 million rows/s., 633.45 MB/s.)


select reporttime,appid,uid,clientversioncode,sdkversioncode,starttimestamp,backgroundtotal,foregroundtotal,firstiframesize,firstiframedecodetime,extras,mclientip,mnc,mcc,vsipsuccess,msipsuccess,vsipfail,msipfail,mediaflag,videomutetype,owneruid from file('./bigolive_audience_stats_orc.orc', ORC)    Format Null; 

0 rows in set. Elapsed: 0.433 sec. Processed 1.20 million rows, 227.43 MB (2.78 million rows/s., 525.35 MB/s.)
0 rows in set. Elapsed: 0.469 sec. Processed 1.20 million rows, 227.43 MB (2.57 million rows/s., 484.86 MB/s.)
0 rows in set. Elapsed: 0.426 sec. Processed 1.20 million rows, 227.43 MB (2.83 million rows/s., 534.22 MB/s.)

Q2: read mixed int/string type columns, speed up by 1.6x

select * EXCEPT (logininfolist)  from file('./bigolive_audience_stats_orc.orc', ORC)  format Null;
0 rows in set. Elapsed: 5.687 sec. Processed 1.20 million rows, 2.46 GB (211.59 thousand rows/s., 431.98 MB/s.)
0 rows in set. Elapsed: 6.001 sec. Processed 1.20 million rows, 2.46 GB (200.51 thousand rows/s., 409.36 MB/s.)
0 rows in set. Elapsed: 5.369 sec. Processed 1.20 million rows, 2.46 GB (224.12 thousand rows/s., 457.56 MB/s.)

select * EXCEPT (logininfolist)  from file('./bigolive_audience_stats_orc.orc', NativeORC)  format Null;
0 rows in set. Elapsed: 3.512 sec. Processed 1.20 million rows, 2.46 GB (342.67 thousand rows/s., 699.58 MB/s.)
0 rows in set. Elapsed: 3.533 sec. Processed 1.20 million rows, 2.46 GB (340.58 thousand rows/s., 695.32 MB/s.)
0 rows in set. Elapsed: 3.664 sec. Processed 1.20 million rows, 2.46 GB (328.43 thousand rows/s., 670.52 MB/s.)

@taiyang-li taiyang-li marked this pull request as draft August 11, 2023 10:53
@robot-ch-test-poll1 robot-ch-test-poll1 added the pr-improvement Pull request with some product improvements label Aug 11, 2023
@robot-ch-test-poll1
Copy link
Copy Markdown
Contributor

robot-ch-test-poll1 commented Aug 11, 2023 •
edited
Loading

This is an automated comment for commit e5a399d with description of existing statuses. It's updated for the latest CI running
The full report is available here
The overall status of the commit is 🔴 failure

Check nameDescriptionStatus
AST fuzzerRuns randomly generated queries to catch program errors. The build type is optionally given in parenthesis. If it fails, ask a maintainer for help🟢 success
CI runningA meta-check that indicates the running CI. Normally, it's in success or pending state. The failed status indicates some problems with the PR🟡 pending
ClickHouse build checkBuilds ClickHouse in various configurations for use in further steps. You have to fix the builds that fail. Build logs often has enough information to fix the error, but you might have to reproduce the failure locally. The cmake options can be found in the build log, grepping for cmake. Use these options and follow the general build process🟢 success
Compatibility checkChecks that clickhouse binary runs on distributions with old libc versions. If it fails, ask a maintainer for help🟢 success
Docker image for serversThe check to build and optionally push the mentioned image to docker hub🟢 success
Fast testNormally this is the first check that is ran for a PR. It builds ClickHouse and runs most of stateless functional tests, omitting some. If it fails, further checks are not started until it is fixed. Look at the report to see which tests fail, then reproduce the failure locally as described here🟢 success
Flaky testsChecks if new added or modified tests are flaky by running them repeatedly, in parallel, with more randomization. Functional tests are run 100 times with address sanitizer, and additional randomization of thread scheduling. Integrational tests are run up to 10 times. If at least once a new test has failed, or was too long, this check will be red. We don't allow flaky tests, read the doc🟢 success
Install packagesChecks that the built packages are installable in a clear environment🟢 success
Integration testsThe integration tests report. In parenthesis the package type is given, and in square brackets are the optional part/total tests🔴 failure
Mergeable CheckChecks if all other necessary checks are successful🟢 success
Performance ComparisonMeasure changes in query performance. The performance test report is described in detail here. In square brackets are the optional part/total tests🟢 success
Push to DockerhubThe check for building and pushing the CI related docker images to docker hub🟢 success
SQLTestThere's no description for the check yet, please add it to tests/ci/ci_config.py:CHECK_DESCRIPTIONS🟢 success
SQLancerFuzzing tests that detect logical bugs with SQLancer tool🟢 success
SqllogicRun clickhouse on the sqllogic test set against sqlite and checks that all statements are passed🟢 success
Stateful testsRuns stateful functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc🟢 success
Stateless testsRuns stateless functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc🟢 success
Stress testRuns stateless functional tests concurrently from several clients to detect concurrency-related errors🟢 success
Style CheckRuns a set of checks to keep the code style clean. If some of tests failed, see the related log from the report🟢 success
Unit testsRuns the unit tests for different release types🟢 success
Upgrade checkRuns stress tests on server version from last release and then tries to upgrade it to the version from the PR. It checks if the new server can successfully startup without any errors, crashes or sanitizer asserts🟢 success

@alexey-milovidov alexey-milovidov added the can be tested Allows running workflows for external contributors label Aug 13, 2023
@taiyang-li taiyang-li changed the title [WIP] implement native orc input format without arrow to improve performance Implement native orc input format without arrow to improve performance Aug 14, 2023
@taiyang-li taiyang-li marked this pull request as ready for review August 14, 2023 07:57
@taiyang-li taiyang-li marked this pull request as draft August 14, 2023 07:57
@taiyang-li taiyang-li marked this pull request as ready for review August 14, 2023 09:08
@taiyang-li taiyang-li mentioned this pull request Aug 14, 2023
Merged
@taiyang-li
Copy link
Copy Markdown
Contributor Author

taiyang-li commented Aug 15, 2023 •
edited
Loading

@alexey-milovidov could you help add a label pr-performance to trigger performance comparision ?

@alexey-milovidov alexey-milovidov added the pr-performance Pull request with some performance improvements label Aug 15, 2023
@alexey-milovidov
Copy link
Copy Markdown
Member

alexey-milovidov commented Aug 15, 2023

Ok.

@alexey-milovidov
Copy link
Copy Markdown
Member

alexey-milovidov commented Aug 15, 2023

Just in case - performance comparison for x86_64 is run for every PR, only AArch64 comparison requires pr-performance.

@robot-ch-test-poll robot-ch-test-poll added submodule changed At least one submodule changed in this PR. and removed pr-improvement Pull request with some product improvements labels Aug 15, 2023
@taiyang-li taiyang-li mentioned this pull request Aug 15, 2023
Closed
@taiyang-li
Copy link
Copy Markdown
Contributor Author

taiyang-li commented Aug 16, 2023 •
edited
Loading

The performance report proves the benefits of native orc input format

image

image

@al13n321 al13n321 self-assigned this Aug 17, 2023
al13n321
al13n321 requested changes Aug 18, 2023
View reviewed changes
Copy link
Copy Markdown
Member

@al13n321 al13n321 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good, just some minor comments.

@taiyang-li
Copy link
Copy Markdown
Contributor Author

taiyang-li commented Aug 21, 2023

@al13n321 I have finished requested changes, can you review again? Thanks!

al13n321
al13n321 approved these changes Aug 21, 2023
View reviewed changes
Copy link
Copy Markdown
Member

@al13n321 al13n321 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thanks for the contribution!

@al13n321 al13n321 merged commit 6009e1b into ClickHouse:master Aug 21, 2023
@baibaichen baibaichen mentioned this pull request Aug 22, 2023
Merged
@taiyang-li taiyang-li mentioned this pull request Sep 12, 2023
Merged
yokofly pushed a commit to timeplus-io/proton that referenced this pull request Mar 1, 2025
@yokofly
Upgrade Protobuf (#6111)
d03e5b1 
* Bump protobuf to v3.18.x (ref: ClickHouse/ClickHouse#50400)

* Bump protobuf to v3.19.6 (ref: ClickHouse/ClickHouse#55963)

* Bump protobuf to v3.19.6 (ref: ClickHouse/ClickHouse#50920)

* Update contribe/orc (ref: ClickHouse/ClickHouse#53324)

* Bump protobuf to v21.9 (ref: ClickHouse/ClickHouse#56014)

* Removed unused code in google-protobuf-cmake/CMakeLists.txt (ref: ClickHouse/ClickHouse#56102)

* Bump gRPC to v1.55.4 and protobuf to v22.5 (ref: ClickHouse/ClickHouse#56664)

* Bump gRPC to v1.59 and protobuf to v24.4 (ref: ClickHouse/ClickHouse#56853)

* enable functions in readonly mode (ref: ClickHouse/ClickHouse#42708)

* Introduce more fields for in-source function documentation (ref: ClickHouse/ClickHouse#49300)

* Add ability to convert timeplusd table structure to CapnProto/Protobuf format schema (ref: ClickHouse/ClickHouse#52278)

* Well known protos (ref: ClickHouse/ClickHouse#56741)

* Update protobuf to v25.1 (ref: ClickHouse/ClickHouse#58020)

* fixed unit_tests_dbms build

* Avoid ignoring errors of execute_process() (set COMMAND_ERROR_IS_FATAL=ANY) (ref: ClickHouse/ClickHouse#68267)

* fixed reigsterFunction

* fixed error after rebase
yokofly added a commit to timeplus-io/proton that referenced this pull request Mar 1, 2025
@yokofly
Upgrade Protobuf (#6111)
900634d 
* Bump protobuf to v3.18.x (ref: ClickHouse/ClickHouse#50400)

* Bump protobuf to v3.19.6 (ref: ClickHouse/ClickHouse#55963)

* Bump protobuf to v3.19.6 (ref: ClickHouse/ClickHouse#50920)

* Update contribe/orc (ref: ClickHouse/ClickHouse#53324)

* Bump protobuf to v21.9 (ref: ClickHouse/ClickHouse#56014)

* Removed unused code in google-protobuf-cmake/CMakeLists.txt (ref: ClickHouse/ClickHouse#56102)

* Bump gRPC to v1.55.4 and protobuf to v22.5 (ref: ClickHouse/ClickHouse#56664)

* Bump gRPC to v1.59 and protobuf to v24.4 (ref: ClickHouse/ClickHouse#56853)

* enable functions in readonly mode (ref: ClickHouse/ClickHouse#42708)

* Introduce more fields for in-source function documentation (ref: ClickHouse/ClickHouse#49300)

* Add ability to convert timeplusd table structure to CapnProto/Protobuf format schema (ref: ClickHouse/ClickHouse#52278)

* Well known protos (ref: ClickHouse/ClickHouse#56741)

* Update protobuf to v25.1 (ref: ClickHouse/ClickHouse#58020)

* fixed unit_tests_dbms build

* Avoid ignoring errors of execute_process() (set COMMAND_ERROR_IS_FATAL=ANY) (ref: ClickHouse/ClickHouse#68267)

* fixed reigsterFunction

* fixed error after rebase
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@al13n321 al13n321 al13n321 approved these changes

Assignees

@al13n321 al13n321

Labels

can be tested Allows running workflows for external contributors pr-performance Pull request with some performance improvements submodule changed At least one submodule changed in this PR.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

5 participants

@taiyang-li @robot-ch-test-poll1 @alexey-milovidov @al13n321 @robot-ch-test-poll